# Integration design worksheet

Fictional examples only. Keep credentials and sensitive records out of shared copies.
This is a planning document, not an IFS import file or a guarantee that a mechanism is available in your environment.

## Approach
- Business outcome:
- Who initiates each exchange (IFS or the external system):
- Synchronous/on-demand or batch/scheduled:
- Chosen mechanism (projection API / workflow / IFS Connect / file exchange / custom projection):
- Why the alternatives were rejected:

## Discovery
- Candidate projection and entity set:
- Key fields and required properties:
- Supported operations and business actions:
- API policy, limits, and ownership:

## Authentication and access
- Integration identity (dedicated, not personal/shared):
- Client credential owner and rotation plan:
- Granted permission sets and why each is needed:
- Token caching and reuse approach:

## Reading data
- Selected fields:
- Filter conditions:
- Ordering and tie-breaker:
- Pagination/continuation approach:

## Writing data
- Create, update, or business action:
- Required vs. optional fields confirmed against the contract:
- ETag/concurrency handling for updates:
- Idempotency/duplicate-prevention approach:

## Reliability and operations
- Correlation identifier strategy:
- Retryable vs. non-retryable failure classification:
- Reconciliation approach for unknown outcomes:
- Monitoring/alerting owner:

## Verification
- Test matrix and actual evidence:
- Real call tested under intended integration identity:
- Denied/insufficient-scope case checked:
- Target-environment endpoint and grant types confirmed:
- Reviewer / date / verified release: